Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
DOCUMENT RESUME
ED 042 515 PS 003 456
AUTHORTITLE
INSTITUTION
SPONS AGENCYPUB DATENOTE
EDRS PRICEDESCRIPTORS
IDENTIFIERS
Herman, Hannah; Adkins, Dorothy C.Hawaii Head Start Evaluation Follow-Up--1968-69.Fina 1 Report.Hawaii Univ., Honolulu. Head Start Research andEvaluation Center.Office of Economic Opportunity, Washington, D.C.Jan 7047p.
EDRS Price MF-$0.25 HC-$2.45*Achievement Gains, Behavioral Objectives, BehaviorProblems, *Cognitive Development, CognitiveObjectives, Creative Activities, *IntelligenceQuotient, Measurement, Perceptual Motor Learning,*Preschool Programs, *Program Evaluation,Socioeconomic Status, Summer Programs, TeacherAttitudes, Verbal CommunicationGumpgookies, *Head Start, Preschool Inventory,Wechsler Preschool Primary Scale Of Intelligence,WPPSI
ABSTRACTThis study compared the performance of kindergarten
children who had participated in a full-year Head Start program(FYHS) with that of children who had attended summer Head Start(SHS). FYHS subjects at each of two elementary schools were selectedat random from eligible applicants. SHS comparison groups were madeup primarily of children from the same initial lists of subjects. Ona group of measures administered about 8 months after completion ofthe Head Start programs, no significant differences were shownbetween FHYS and SHS children, or between the two elementary schoolgroups. SHS children, however, earned significantly higher IQ scoresat the end of kindergarten than they had early in Head Start, andFYHS children showed a significant progressive increase in IQ over a2-year period covering Head Start and kindergarten. This continuousincrease in IQ is noteworthy, since previous studies have found aleveling-off effect following an initial gain in Head Start. It isrecommended that follow-up studies concentrate on the long-rangeeffects of FYHS programs with defined curricula focused on cognitivedevelopment. (Author/NH)
U. DEPARTMENT OF REALM, EDUCA1TON WarDREOFFICE OF EDUCATION
THIS DOCUMENT HAS BEEN REPRODUCED EXACTLY AS RECEIVED FROM THEur% PERSON OR ORGANIZATION ORIGINATING IT. POINIS OF VIEW OR OPINIONS
STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDUCATIONr4POSITION OR POLICY.
ur% University of Hawaii(NI Read Start Evaluation and Research Center
Dorothy C. Adkins, Director
c:3
1.0 Final Report
Hawaii Head Start Evaluation Follow-Up--1968-69
Hannah Herman, Assistant ResearcherDorothy C. Adkins, Researcher
The research reported herein was performed pursuant toa contract with the Office of Economic Opportunity,Executive Office of the President, Washington, D. C.
20506. The opinions expressed herein are those of theauthors and should not he construed as representingthe opinions or policy of any agency of the United
States Government.
Contract No. 0E0 4121
Education 'Research and Development CenterDavid G. Ryans, Director
College of EducationUniversity of HawaiiHonolulu, Hawaii
January, 1970
F-OE6 It-
TABLE O1 CONTENTS
List of Tables
Foreword Iii
Introduction
Procedure
1
2
Pull-Year Head Start Programs 2
Summer Head Start Programs 5
Description of the Elementary Schools 6
Sample and Experimental Desien 7
h.esults and Discussion 14
Teacher Interview and Checklist 15
Analysis of Test Data 21
WPPSI Results 25
Results of PSI, GumpRookies, Inventory of Factors AffectingWPPSI Performance and "Work" Per cent 30
Longitudinal Analysis of NPPSI Performance IQ's 33
Conclusions and Recommendations 37
Appendix 40
References 42
LIST OF TABLES
PROCEDURE
Table 1--Distribution of Children According to Amount ofHead Start Experience and Assignment to KindergartenClass 8
Table 2--Experimental Design 10
Table 3--Comparison of the Family Structure of FYHS andSHS Children 11
Table 4--Comparison of the Educational Level of the Fathers ofFYHS and SHS Children 11
Table 5--Comparison of the Educational Level of the Mothers ofFYHS and SHS Children 12
Table 6--Comparison of the Occupations of the Fathers ofFYHS and SHS Children 12
RESULTS AND DISCUSSION
Table 7 -- Teachers' Reports of Severe Behavior Problems 16
Table 8-- Teachers' Reports of Mild Behavior Problems 17
Table 9- -Child Behaviors That Please Teachers 19
Table 10--Program Focus of Kindergarten Classes 20
Table 11--Educational Goals of Kindergarten Classes 22
Table 12--Self-Concepts of Kindergarten Teachers 23
Table 13--Means and Standard Deviations for WPPSI Verbal Scaleand Verbal IQ (Spring 1969) 26
Table 14-- Analysis of Covariance for WETS' Verbal Scale 27
Table 15--Means and Standard Deviations for WPPSI PerformanceSubtests, Performance IQ, and Full-Scale IQ
(Spring 1969) 28
Table 16--Analysis of Covariance for T.IPPSI Performance Scale
and UPPSI Full-Scale IQ 29
Table 17--Means and Standard Deviations for Four ResponseMeasures (Spring 1969) 31
List of Tables (cont.)
Table 18--Summary of Analysis of Covariance (PreschoolInventor and Gurwookies) and Analysis of Variance( WPPSI "Work"% and Inventory of Factors Affecting,Test Performance) 32
Table 19--Mean WPPSI Performance IQ's for Successive Yearsfor FYHS and SHS Children 34
Table 20--Analysis of Variance of WPPSI Performance 10'sfor FYRS Children for Three Successive Years 34
Table 21--Analysis of Variance of VIPPSI Performance IQ's forTwo Successive Years 35
APPENDIX -- Inter - correlations Among Follow-Up Measures andPrevious Intelligence Test Data 40
ii
FOREWORD
This study was conducted with the cooperation of the following
personnel at the two sample schools.
Maili Elementary School:
Mr. Shigeo Kimura, principal.Mrs. Ramona Souza, teacher of full-year Head Start, 1967-68.Mrs. Sally Dockham, teacher of summer Head Start, 1968.Mrs. Sally Dockham, teacher of kindergarten, 1968-09.Miss Carol Hiraoka, teacher of kindergarten, 1968-69.Mrs. Eleanor Kama, teacher of kindergarten, 1968-69.Mrs. Marjorie Overton, teacher of kindergarten, 1968-69.Mrs. Eva Yonemori, teacher of kindergarten, 1968-69.Miss Marion Ueda, teacher of kindergarten, 1968-69.
Makaha Elementary School:
Mr. Sam Oshio, acting principal.Mrs. Karie Bryant, teacher of full-year Head Start, 1967-68.Miss Millicent Fo, teacher of kindergarten.Miss Ellen Horiuchi, teacher of kindergarten.Mrs. Ramona Souza, teacher of kindergarten.Miss Tony De Terlizzi, teacher of kindergarten.Miss Vivian Yonehara, teacher of kindergarten.
Testing and collection of data were done by the following personnel
of the Head Start Research Center staff: Betty Crooker, Carole Hodges,
Virginia Lerner, Annette Okimoto, Ann Pavelko, Katherine Payton,
Betty Parker, Betty Elrod, and Annie Worth.
iii
A series of studies summarized in a review of Head Start research
(Grotberg, 1969) indicate that gains made in Head Start on cognitive
measures tend to maintain themselves but that comparable gains ere
usually made by non-Head Start children in elementary school. Follow-up
studies of Head Start after one or two years of traditional school expe-
rience, therefore, have generally failed to find significant differences
between Head Start and non-Head Start children. The non-Head Start com-
parison groups have primarily been subjects matched on a variety of
indices such as age, sex, race, and socioeconomic status, although few
studies have taken all these variables into account. Difficulties in
assuring comparability of subjects when post hoc matching procedures
are used, however, are considerable. Since in two communities on Oahu,
Hawaii, children who attended full-year Head Start (FYHS) were randomly
selected from a list of eligible applicants whose parents had applied
for admission to the neighborhood program, it seemed worthwhile to com-
pare the performance in kindergarten of the children who had attended
FYHS with those who had not. Many more summer than full-year programs
had been offered in both areas. Therefore, those children not selected
for a full-year program participated in summer Head Start (SHS), as did
all but three of the FYHS children. The purpose of this study, then,
was to compare the kindergarten performance of children who attended
full-year in addition to summer Head Start with those who attended a
summer program only.
1
PROCEDURE
All the children in the study were drawn from two Head Start Centers
and attended kindergarten at two elementary schools. Descriptions of the
Head Start programs and of the schools follow.
Full-Year Head Start Programs
The two Head Start classes attended by the FYHS children were in-
cluded in the University of Hawaii 1967.68 national evaluation sample.
Both classes were under the jurisdiction of the Hawaii Department of
Education and were affiliated with a particular elementary school. Cen-
ter # 1 was located on the grounds of the elementary school with which
it was affiliated; the second Center was housed in a church in the neigh-
borhood of the other elementary school. Supervision of the programs was
primarily the responsibility of the school principals. Since data col-
lected for the 1967-68 national evaluation included classroom observations,
brief descriptions of the two Head Start programs are provided. The clas-
ses are compared with each other as well as with the combined average
results for the seven classes included in the Hawaii national evaluation
study sample.
Classroom observations were carried out five times during 1967-68,
using an experimental procedure for describing classroom curricula (Obser-
vation of Substantive Curricular Input . Each observation period lasted
two hours, during which a record of the classroom activities was made for
consecutive three-minute segments. The activities that were observed oan
be classified into four categories: 11 perceptual-motor, 2) cognitive,
3) creative, and 4) social and verbal communication.
2
Activities that emphasized primarily perceptual-motor skills were
observed 475 times on the average, 467 times in Center # 1, and 548 times
in Center # 2. The major differences between the two Centers in this
general area of perceptual and sensory tasks were in the relatively great-
er emphasis on outdoor activities and on the use of puzzles and other
games involving visual discrimination in Center # 2. Auditory discri-
mination tasks were observed more frequently in Center # 1. Cognitive
activities, specifically those focused on language development, occurred
more frequently in Center # 2. Language was coded 89 times in Center # 2
and 50 times in Center # 1; the sample average was 71. Within the crea-
tive area, dramatic play occurred much more frequently in Center # 2
(118 times compared with 19), and art activities much more often in
Center # 1 (103 times compared with 39). The figures for the sample
average were 62 and 92, respectively. All types of communication were ob-
served with greater frequency in Center # 2. The numbers of times general
discussions, social communications, or presentations of rules were recor-
ded in Center # 2 were 269, 279, and 56,respectively. The comparable
figures for Center # 1 were 137, 191, and 14. In all cases, the figures for
Center # 2 were higher than the sample average and those for Center # 1
were lower.
Activities that did not fit into any prescribed category were also
noted. Included in this category were such diverse behaviors as fighting,
wandering, and apparently not doing anything. These types of behaviors
were observed 49 times in Center # 1, 126 times in Center # 2, and 92
times on the average. The children in Center # 2 were also more fre-
quently observed alone (362 times) than were those in Center # 1 (234
times); the sample average was 271. Many of these individual child-
observations describe aimless wandering.
3
In summary, then, more activities were observed in Center # 2 than
in Center # 1 in perceptual-motor, cognitive, and communication areas.
Of the creative activities, dramatic play was more frequent in Center # 1,
but general arts and crafts activities were seen much more often in Cen-
ter # 2. There were also more children observed alone and without direc-
tion in Center # 2 than in Center # 1.
Another experimental procedure (Social Interaction Observation) was
used in the 1967-68 national Head Start evaluation to describe the amounts
and types of social interactions that occur between children and between
children and adults. These observations were carried out near the begin-
ning and end of the school year. Analysis of these data reveals that for
both sets of observations, children in Center # 2 had a higher number of
interactions with each other and with adults than did those in Center # 1.
The pre- and post-observation figures for Center # 2 for the mean number
of interactions per child. were 69.8 and 107.4; comparable pre-post fig-
ures for Center # 1 were 55.3 and.73:8. In Center # 2, approximately
75% of child initiations, i.e., attempts to make verbal or nonverbal con-
tact with another child or adult, were responded to both at the begin-
ning and at the end of the year. In Center # 1, 517, of initiations to
peers were responded to initially; 637. received a response during the
post-observations. Approximately 65% of the attempts of the Center # 1
children to make contact with an adult received a response during both
observation periods. In general, there was more socialization among
children and between children and adults in Center # 2 than in Center # 1.
No direct teacher observations were carried out as part of the 1967-
68 national Head Start evaluation. The head teacher at Center # 1 was
a certified elementary school teacher experienced with both preschool and
elementary school youngsters. She was, however, ill during the early
4
months of the program and was completely absent from the class during a
period of approximately five weeks, during which some of the classroom
observations were conducted. The head teacher in Center # 2 was a recent
college graduate without previous teaching experience.
Summer Head Start ro rams involvin FYHS children. The children
from Center # 1 who attended an eight-week summer program in addition to
the full-year program were placed in one class that was taught by their
FYHS teacher. This same teacher taught one of the kindergarten classes
the following year. No systematic observation of the summer program was
made, but the general impression of testers was that there was a much
freer atmosphere than had prevailed during the year. This was confirmed
by the teacher who felt that it was not so necessary to restrict the chil-
dren, since there were no regular classes in session that might be dis-
rupted by excessive noise coming from her classroom.
All the children from Center # 2 who attended FYHS also attended a
summer program. The summer classes were held on the grounds of the ele-
mentary school rather than in the neighborhood church. The FYHS chil-
dren were placed in one class that was taught by a very experienced elemen-
tary school teacher who taught kindergarten during the year at the same school.
Summer Head Start Programs
In Center # 1 all the children in the SHS-only group were in the
same Head Start class. One child included in the Center # 2 SHS-only
group was also in this class. The head teacher was exceptionally well!714
qualified, with extensive training and experience in preschool education.
0 She had served at various times as a consultant and seminar leader in
training programs for Head Start teachers. The SHS-only children from
Center # 2 were distributed across four Head Start classes at the same
5
location, with the exception of the one child noted above, who moved
and entered kindergarten in the school afftliated with Center # 2 and
was therefore included in this group. The summer Head Start teachers
were all certified elementary school teachers, one of whom was the coor-
dinator of the kindergarten program at that school. The summer programs
were eight weeks long, with the usual amount of time devoted to health
and medical care of the children and to general preschool activities.
Observations of these classes were not carried out, but they were pre-
sumed to be fairly typical of the type of summer program conducted in
Hawaii.
Description of the Elementary Schools
The elementary schools at which the Head Start children subsequently
attended kindergarten are referred to as School # 1 and School # 2, to
associate them with the corresponding Head Start Centers. That is,
FYHS and SHS children from Center # 1 attended School # 1 and those from
Center # 2 attended School # 2 during 1968-69.
The two elementary schools are located in the same coastal region
of Oahu, a primarily rural area that has been designated as one of the
two model cities target areas in Hawaii. Both schools serve kindergarten
through the sixth grade in addition to Head Start. School # 2 is slight-
ly larger, having a total of 37 classes including six kindergartens and
approximately 1100 pupils, compared with 34 classes with five kindergar-
tens and approximately 900 pupils at School # 1. The schools were
built at about the same time, but School # 2 has somewhat better facili-
ties and more varied programs. School # 1 does not have a cafeteria,
requiring children to eat in their rooms, and does not have a full-time
nurse, music teacher, or remedial reading specialist. School # 2 has
6
these personnel and in addition provides special classes for children
with intellectual or behavior problems.
In both elementary schools, initial assignment to kindergarten clas-
ses was strictly on the basis of age. In January, the children attending
School # 2 were reassigned to classes on the basis of reading readiness.
The distribution of children at each school in the various kindergarten
classes is presented in Table 1. The major effect of regrouping the
children at School # 2 was to place three FYHS children into the acceler-
ated class CO 5). In both schools the children from FYHS and SHS groups
were fairly widely dispersed across kindergarten classes, and thus dif-
ferences among groups are not likely to be due to particular kindergar-
ten teacher or program variables. Head Start children represented a rela-
tively small proportion of the children in a given class and in some
classes were quite unique.
Sample and Experimental Design
School # 1. From the original pool of Head Start applicants from
this district in the fall of 1967, 16 children who were randomly selected
to attend full-year Head Start comprised the Center # 1 FYHS group. Of
these 16 children, 13 attended a summer Head Start program. The group
consisted of 11 boys and five girls.
Eleven children -- seven boys and four girls -- comprised the Cen-
ter # 1 SHS group. Due to heavy attrition during 1967 among children not
chosen to attend FYHS, it was unfortunately necessary to include in this
group four children who were not in the original pool of subjects but
who attended the same summer Head Start program as did the remaining sub-
jects.
School # 2. Seventeen children -- seven boys and 10 girls -- made
up the FYHS group from School # 2. All of these children were in the ini-
7
TABLE 1
Distribution of Children According to Amountof Head Start Experience and Assignment to Kindergarten Class
School # 1 School # 2
ClassSeptember - MayFYI-IS SHS
September - JanuaryFYHS SHS
February - MayFYHS SHS
1 2 5 3 3 3 3
2 1 2 2 3 1 3
3 7 2 3 0 3 0
4 5 2 5 3 4 3
5 1 0 1 3 4 3
6 - - 3 1 2 1
Totals 16 11 17 13 17 13
8
tial pool of subjects in the fall of 1967 and all attended the same
Head Start class. The SHS comparison group was in the initial pool of
subjects, with the exception of one boy. This boy had attended the same
program as the School # 1 SHS children and then switched to School # 2
for kindergarten. Seven boys and six girls made up this group.
A schematic presentation of the design of the study is given in
Table 2. Since complete data were not obtained for all subjects, most
analyses involve fewer cases than comprised the total sample.
Further description of the sample is provided by comparisons be-
tween the FYHS and SHS children on various socioeconomic dimensions.
These data are presented in Tables 3 to 6. The information was obtained
primarily from school records and consequently is often incomplete.
When percentage comparisons are made, the number of cases for whom data
are missing should be noted.
There was a slightly higher percentage of intact families within
the FYHS group, when percentages were based on the total number of cases
in each group. When the percentage base was limited to the number of
subjects for whom data were available, the percentages of intact families
in FYHS and SHS groups were essentially the same (approximately 807.).
The fathers of children in the FYHS group had had, on the average, some-
what more education than those of the SHS children. Higher percentages
of them were high school graduates or had completed at least nine years
of school. In both FYHS and SHS groups, mothers had more education than
their husbands. When percentages were based on the number of cases for
whom data were available,ahout 50 per cent of the mothers in each group
had completed high school. The unemployment rate of the fathers of both
groups of children was very similar, but the fathers of SHS children had
somewhat higher-level jobs. The fathers of FYHS children were as likely
9
TABLE 2
Experimental Design
School # 1
School# 2
Amount of Head Start Experience
FYHS SHS
16 11
17 13
33 24
10
27
30
57
TABLE 3
Comparison of the Family Structure of FYHS and SHS Children
FamilyStructure
Full-Year Head Start Suer Head St rtSchool # 1 School # 2 Total School # 1 School # 2 Total
N % N % N % N % N N
Unknown Mb tab bO a. OS MOD - - 27.3 3 12.5
Motheronly 25.0 3 17.6 7 21.2 1 9.1 3 23.1 4 16.7
Mother andFather 12 75.0 14 82.3 26 78.8 7 63.6 10 76.9 17 70.1
TABLE 4
Comparison of the Educational Level of the Fathersof FYHS and SHS Children
EducationalLevel
Full-Year Head Start Summer Head StartSchool # 1N %
School # 2N 7.
TotalANT %
School # 1N 7.
School # 2N 7.
--Total
N %
Unknown 4 25.0 3 17.6 7 21.2 5 45.4 4 30.8 9 37.5
1st - 3rd grade -- -- 2 11.8 2 6.1 I 9.1 0 0.0 1 4.2
4th - 6th grade -- -- 1 5.9 1 3.0 2 18.2 0 0.0 2 8.3
7th - 8th grade 2 12.5 3 17.6 5 15.2 1 9.1 3 23.1 4 16.7
9th - 11th grade 5 31.2 5 29.4 10 30.3 -- -- 5 38.5 5 20.8
High Schoolgraduate 5 31.2 3 17.6 8 24.2 2 18.2 1 7.7 3 12.5
11
TABLE 5
Comparison of the Educational Level of the Mothers ofFYHS and SHS Children
Educational Level.
Full-Year Head Start Summer Head Start
School#1N 7.
School#2N V.
Total .
N %
School#1N %
Scho61#2N 7.
Total
N 7.
Unknown 1 5.9 -- 1 3.0 4 36.4 2 15.4 6 25.0
1st - 3rd grade -- -- -- -- _- -- --
4th - 6th grade -- -- -- -- - -- - -- -- --
7th - 8th grade 2 11.8 -- 2 6.1 1 9.1 1 7.7 2 8.3
9th - 11th grade 8 47.1 6 35.3 14 42.4 3 27.3 3 23.1 6 25.0
High School Grad. 5 31.2 10 58.9 15 45.4 3 27.3 6 46.2 9 37.5
Some College .... 1 5.9 1 3.0 -- -- 1 7.7 1 4.2
TABLE 6
Comparison of the Occupations of the Fathers ofFYHS and SHS Children
Full-Year Head Start Summer Head Start
Occupations School#1N 7.
School#2N %
TotalN 7.
School#1N %
School#2N 7.
TotalN 7.
Unknown 4 25.0 4 23.5 8 24.2 3 21%3 4 30.8 7 29.2
UnemplOyed 6 37.5 2 11.8 8 24.2 2 1b.2 3 23.1 5 20.8
Unskilled Worker 3 18.8 3 17.6 6 18.2 1 9.1 1 7.7 2 8.3
Semiskilled Worker 2 12.5 4 23.5 6 18.2 3 27..3 3 23.1 6 25.0
Skilled Worker 1 6.2 2 11.8 3 9.1 1 9.1 2 15.4 3 12.5
Clerical & Sales -- -- 2 11.8 2 6.1 -- -- - --
Semiprofessional -- -- -- -- -- -. 1 9.1 -- -- 1 4.2
12
to be in unskilled as in semi-skilled jobs, whereas-the SHS fathers
were more often listed in semi-skilled occupations. Only one mother
with a child who had attended FYHS was listed as employed; three mothers
of SHS children were listed am employed on the school records of their
children.
In general, there were no marked differences in the socioeconomic
status of the FYHS and SHS groups as a whole. Most of the families were
intact; most fathers had between nine and 11 years of education and were
either unemployed or in an unskilled or semi-skilled job; very few of
the mothers worked, although approximately half of them were high-school
graduates. Comparison of these figures with data available for the
kindergarten classmates of the Head Start children is of interest.
Eighty-one per cent of a total of 226 kindergarten children from both
schools who did not attend any Head Start program came from intact fami-
lies, a figure essentially the same as that for Head Start children. A
higher percentage of the fathers of these children, however, were high-
school graduates (34.7%) and fewer of them were unemployed (10%). Only
one father was listed as a college graduate, however, and the most fre-
quent types of employment were in semi-skilled (32.4%) or skilled jobs
(21.2%). The educational backgrounds of the mothers were comparable to
those of the mothers of Head Start children, and, similarly, these moth-
ers were infrequently employed. The kindergarten classmates of the Head
Start children were not from radically different socioeconomic back-
grounds; rather, they tended to have fathers with slightly more educa-
tion, who were more likely to be employed.
13
RESULTS AND DISCUSSION
A common battery of tests was administered by all Head Start Evalua-
tion and Research Centers conducting a follow-up study of samples of chil-
dren who had participated in the 1967-68 national Head Start evaluation.
The instruments used in the follow-up studies included the Wechsler Pre-
school and Primer Scale of Intelli ence ( WPPSI); a procedure for evalua-
ting response styles on the WPPSI (a modification of a system developed
by Hertzig, Birch, Thomas, and Mendez, 1968, for use with the Stanford-
Binet); the WPPSI Inventory, a rating scale of behavior during the WPPSI
administration (an adaptation of the Inventor of Factors Affecting the
Stanford-Binet); an experimental measure of achievement motivation, Gump-
Rookies, developed at the University of Hawaii (Adkins and Ballif, Spring
1969 form); a brief teacher interview; and a checklist on educational objec-
tives and emphases completed by the kindergarten teachers. The data reported
for the measure based on the work of Hertzig, et al. (1968), consists of a
"work"%score,the degree of responsiveness to the demands of the WPPSI, i.e.,
attempts to handle the tasks presented rather than engaging in irrelevant
behavior. The WPPSI Inventory is a rating by the examiner of the child's
reactions to the total test situation, with lower scores reflecting more
adaptive behavior. The child's task in the ammoltiel test is to select
one of the two imaginary figures in each of a series of situations which is
most like himself. The test score is the number of times the child chooses
the Gumpgookie whose behavior is assumed to reflect achievement motivation.
In addition to these measures, the 64-item form of the Preschool Inventory
(PSI) was administered to the children in the Hawaii follow-up sample,
14
since no achievement measure was included in the common battery. The PSI,
as well as the other individual follow-up measures with the exception of
the WPPSI, were used in the 1968-69 national Head Start evaluation. A sum-
mary of findings from the teacher interview and checklist is presented first,
followed by the results of the individual testing.
Teacher Interview and Checklist
Five kindergarten teachers from School # 1 and six from School # 2 were
interviewed. Included in the interview were questions related to the teacher's
educational background and experience. All the kindergarten teachers had a
Bachelor's degree in Elementary Education. In addition, two of the five
School # 1 teachers and four of the six teachers from School # 2 had a fifth-
year certificate and one teacher had an M.A. degree In Education. The
teachers at School # 1 were considerably younger and consequently less expe-
rienced than those from School # 2. They had taught for an average of 3.2
years, with an average of 2.2 years at the kindergarten level. The teachers
from School # 2 had an average of 14 years of teaching experience and nine
of these years, on the average, were as kindergarten teachers.
During the interview each teacher was asked what she thought was the
worst thing children in her class did, what was a little thing they did
that she felt they shouldn't, and what was something they did that pleased
her. She was also asked to describe what she did or said when each of these
behaviors occurred. Table 7 summarizes teachers' reports of what they con-
sidered the least desirable behavior of their children and broadly classi-
fies their ways of handling these misbehaviors. Comparable data for mild
infractions are presented in Table 8. Failure to listen was mentioned by
three teachers from School # 2 and by one School # 1 teacher; taking things
that didn't belong to the child was noted by two teachers from School # 2;
15
TABLE 7
Teachers' Reports of Severe Behavior Problems
Severe Infractions .Teacher Reactions:
Not listening
N
IONE.
School#1 N School#2
1 Constructive- 2
1
PunishmentPunishment +Constructive
Taking things 2 Constructive+ Punishment
Fighting 1 Punishment 1 Punishment +Constructive
Bad manners 1 Constructive
Lack of self-control 1 Punishment
No response
Total 5 6
16
TABLE 8
Teachers' Reports of Mild Behavior Problems
Mild Infractions Teacher Reactions
Speaking out of turn
N School0 1 N School# 2
1 Constructive 2 Punishment +Constructive
Running around the room 2 Constructive 1 Punishtent 4-.
Constructive
Tattling 1 Punishment +Constructive
Fighting 1 Punishment
Lack of initiative 1 Constructive
No response 1 1
Total 5 6
17
and the remaining severe problems were cited by individual teachers.
Teacher reactions to these problems that provided the children with a reason-
able explanation for why their behavior was unacceptable, or which suggested
an alternative activity that was acceptable, were considered constructive.
All other types of responses were classified as punishments. It can be seen
from Table 7 that punishment responses to severe infractions were given by
more teachers from both schools than were constructive responses. The re-
verse was true, however, for behaviors that were considered less disruptive
by the teachers (Table 8). All teachers in School # 1 who responded to
this item gave reactions that were classified as constructive, and only one
teacher in School # 2 gave a strictly punitive response. Apparently, the
teachers were better able to handle problems that they classified as rela-
tively minor than those they considered severe.
The uniform reaction reported by all teachers when the children did
something that pleased them was to provide verbal praise (Table 9). Activ-
ities that reflect initiative were the most frequently mentioned desirable
behaviors by teachers from School # 2; responses of the School lb 1 teachers
were more diverse.
The checklist of educational objectives was completed independently by
the teachers--one teacher from School lb 2 failed to return the form. From
the list of program focuses presented in Table 10, each teacher marked the
three most descriptive of her class. Eight of the 10 teachers responding
to this question selected "social experience" as oni f the primary focuses
of their programs. Most of the teachers at School # 1 also checked language
and the "whole child" as primary program orientations. School # 2 teachers
tended to emphasize the development of the child's self-concept. The empha-
ses of the kindergarten programs apparently were primarily in social-
18
TABLE 9
Child Behaviors That Please Teachers
INIMMmos
Child Behaviors Teacher Reactions
School # 1 Cchool # 2
Activities done through chil-dren's initiative and indepen-dence
1 Verbal praise Verbal praiscNo response
Sharing things 2 Verbal praise
Verbalizing problem ratherthan fighting
1 Verbal praise
Good attitude 1 Verbal praise
Courteous 1 Verbal praise
No response
Total 5 6
39
TABLE 10
Program Focus of Kindergarten Classes
School# 1 School# 2 Total----
1. Parent-centered 1 l 1
.....--2. Child-centered 1 1 2
- ........
3. Family-centered
4. Teacher-centered
5. Material-centered
6. Task-oriented 1 1
7. Mental health-oriented 1 1
8. Language-oriented 2 4 6
9. Social experience-oriented 4 4 8
10. Concept-oriented------------ 2 2
11. Academically oriented-----12. Reading-oriented
13. Self-concept-oriented-....--
14. The "whole child " - oriented
15. Other (Specify)
Total 15 15 30
Note.--Each teacher chose three focuses.
20
emotional development rather than in cognitive or academic areas. Some
further evidence of this is seen in Table 11. When asked to select five
educational goals from a list of 40, the most frequent choice was "working
and playing cooperatively." The other goals mentioned by several teachers
from both schools were for children to learn to follow directions and to
feel secure in school. School # 1 teachers also emphasized the development
of self-confidence and the ability of the child to use his knowledge more
effectively. Although the focuses on following directions and on adap-
tiveuse of information suggest some concern with cognitive functioning, in
gLneral the goals chosen reflect an emphasis on the affective domain. No
teacher, for example, checked reliance on verbal communication rather than
on gestur:ce, or the development of problem-solving abilities as goals in her
curriculum.
The teachers were also asked to check the teacher-roles most descriptive
of themselves from the list of alternatives presented in Table 12. All
teachers described themselves as "motivators" and one-half as "transmitters
of knowledge and skills." Those from School # 1 also tended to define their
roles as designers of learning experiences and as arrangers of reinforcement
contingencies. These choices undoubtedly reflect their relatively recent
training in comparison with the teachers from School # 2.
Analysis of Test Data
Children in both FYHS groups were administered a variety of tests, in-
cluding the WPPSI Performance Scale and Stanford-Binet, in the fall of 1967
and again in the spring of 1968 in accordance with the requirements of the
1967-68 national Head Start evaluation. The WPPSI Performance Scale was
also included in the follow-up battery administered to these children in the
spring of 1969. Thus, intelligence test data for most of the FYHS children
21
TABLE 11
Educational Goals of Kindergarten Classes
School#1 School# 2 Total
1. Partici.ation in :rou. activities2. Trust of adults 1 1
3. Familiarity with books, paper,crayons, pencils, etc.
1
4. Observing safety habits5. Gain to the toilet alone6. Tidiness 1
.
1
7. Handlin: books carefullB. En 'o in: stories 29. Standin: u for his own ri:hts
10. Reading11. S.eakin: more12. Solvin roblems13. Usin: what he knows more effectivel14. Speaking clearl 1 1
15. Thinkin: lo:icall16. Identifying cause-effect relation-
shis17. Enipying other children18. Acce.tin new .eo.le without fear19. Taking turns .
320. Feelin: secure in a school situation 4 1. 521. Caring for and.picking u. materials i 122. Following directions 4 323. Putting on and taking off his ownwras24. Completing a task before starting
another25. Observin: :ood health .ractices26. Relying on verbal communication
more than on gesture27. Workin: and .1a in: coo.erativel 3 5 8
.
28. Res.ectin the ri:hts of others 1 329. Sharing ideas and materials 130. Using good table manners31. Working independently32. Leadin: effectivel,33. Followin: effectivel34. Acce.tin: :rou decisions35. Ex.ressin his neative feelin:s
2§-1-Er"81-.M.-.L81-.-2-.---tivefeelin37. Being confident of himself 4 238. Acceptiv: authorit39. Showing mastery of quantitative
concets and o.eratlons40. Other (Specify) 1
Total 25 25 50
Note.--Each teacher chose five goals.
22
TABLE 12
Self-Concepts of Kindergarten Teachers
Schooltl School #k Total
1. Tranamitter of knowledge andskills
2 3 5----
2. Motivator 5 5 103. Problem solver4. Transmitter of culture5. Model..6f.bahavior 1 1
6. Hypothesis tester or expert..mentalist
.
7. Classroom manaer 1 2 3
LitAgentoichaue9. Profesdional specialist
10. Socializing agent 1 1 211. Diagnostician12. General rofessional 1 113. Designer of learning and ex-
perience3 1 4
14. Devel.ier of human otential 1 2 315. Group2rocesses s.ecialist16. Arranger of reinforcement contin-
gencies or success experiences3 1
2
4
217. Transmitter of moral standardsor values,
18. Developer of knowledge andskills
1 1 2
--__,
19. Administrator,----
20. Pupil/parent advisor, counse177--w21. Observer 1 122. Demonstrator23. Record kee er 1
----1
1
--.
24. Anal st (behavior, achievement,etc.
1
25. Other (Specify) 1
2 -----261
Total
--_
40
----
Note.--Each teacher chose four concepts.
23
were available near the beginning of Head Start, near the end of the full-
year program, and one year later, towards the end of kindergarten. Most of
the FYHS children also attended Head Start in the summer of 1968, but they
were not tested following the summer program.
Children in the SHS groups were given the WPPSI Performance Scale in
July, 1968, in the early weeks of summer Head Start. Some of these children
had also taken the Stanford-Binet in the spring of 1968. The SHS groups
were not retested at the end of the summer program but were administered
the follow-up battery in the spring of 1969 at the same time as the FYHS
groups. In comparisons of the kindergarten performance of children exposed
to a full-year Head Start program with those who attended a summer program
only, it was desirable to use previous data where appropriate as a covariate
in the analyses, The most relevant and complete data across subjects was
the WPPSI Performance IQ obtained prior to extensive exposure to Head Start
(i.e., in the fall of 1967 for FIRS and in the summer of 1968 for SHS
children). These data were not complete, however, due to untestability of
some subjects, absences, and invalid testing by one examiner. In order not
to reduce the small sample further, for five subjects in FYHS School # 1
and one subject in SHS School # 2, for which appropriate Binet but not WPPSI
data were available, the Binet IQ's were used in the covariate analyses.
In view of the fairly high correlations between the tests (see WPPSI Manual,
p. 34, and Appendix) this substitution seemed justifiable. In order to
make the IQ data even more comparable, all IQ's were converted to Z scores
using standard deviations of 15 and 16 for the WPPSI and Stanford - Binet
scores, respectively. The means of the covariates when 100 was added to
each Z score were 99.43 and 99.12 for the School # 1 FIRS and SHS groups,
respectively, and 99.43 and 99.23 for the comparable School # 2 groups.
24
WPPSI Results
The means and standard deviations for the Verbal subtests of the WPPSI
for FYHS and SHS groups administered near the end of kindergarten are pre-
sented in Table 13. The analysis of covariance for these subtesta based on
the adjusted means is given in Table 14. No significant differences were
found between FYHS and SHS children or between children attending one or the
other of the two schools on any yerbal subtests or on total Verbal IQ. Addi-
tionally, there were no significant interactions between length of Head Start
participation and attendance at either school for any of the verbal measures.
Table 15 provides the means and standard deviations for the WPPSI
Performance subtests and the overall WPPSI IQ. The analysis of covariance
for these data is summarized in Table 16. As with the Verbal subtesta,
there were no significant results for any of these measures. It is of
interest, however, tc, note the markedly superior functioning of all groups
of children on the Performance subtests in comparison with the Verbal sub-
tests. The magnitude of the differences suggests that previous exposure to
the Performance subtests would be a highly unlikely explanation for these
findings. On Performance sutests the children on the average earned scores
clearly within the normal range, whereas on the Verbal siubtests they tended
tc perform at one standard deviation or more below the mean for their ages.
Within the Performance Scale the groups consistently did best on the two
tasks requiring the use of a pencil, in one case to copy geometric figures
and in the other to draw a path from the stetting point to the exit of a
series of mazes. Within the Verbal Scale, the Arithmetic test produced the
highest scores and the Vocabulary subtest the lowest. It would be cd inter-
est to investigate whether these findings for specific subtests are related
to program emphases or activities within these kindergarten curricula or
25
TABLE 13
Means and Standard Deviations forWPPS/ Verbal Scale and Verbal IQ
(Spring 1969)
Test Group N Mean SD
Information School 1FYHS 14 7.43 2.21
SHS 7 7.43 2.94
School 2FYHS 14 6.93 2.20
SHS 13 8.23 3.19
Comprehension School 1FYHS 14 6.21 2.91
SHS 7 7.00 2.71
School 2FYHS Li 6.64 3.25
SHS 13 7.31 2.18
Arithmetic School 1FYHS 14 8.79 1.97
SHS 7 9.43 2.44
School 2FYHS 14 8.43 2.80
SHS 13 8.23 2.31
Similarities School 1FYHS 14 6.71 3.79
SHS 7 7.00 2.94
School 2FYHS 14 7.43 3.52
SHS 13 7.54 2.44
Vocabulary School 1FYHS 14 6.07 2.56
SHS 7 6.14 1.21
School 2FYHS 14 5.86 3.23
SHS 13 7.00 3.03
WPPSI Verbal IQ School 1FYHS 14 82.43 11.65
SHS 7 83.71 13.01
School 2FYHS 14 81.50 13.59
SHS 13 85.38 11.13
26
TABLE 14
Analysis of Covariance for WPPSI Verbal Scale
Telt Source df Adjusted SS MS
1
F
WPPSI VerbalSubtests
Information A (Schools) 1 .06 .06 .01
B (FYHS vs. SHS) 1 11.25 11.25 1.95AB 1 3.72 3.72 .64
Error 43 248.28 5.77
Comprehension A (Schools) 1 .98 .98 .14
B (FYHS vs. SHS) 1 11.81 11.81 1.64AB 1 .16 .16 .02
Error 43 308.98 7.18
Arithmetic A (Schools) 1
...,
7.89 7.89 1.58B (FYHS vs. SHS) 1 3.10 3.10 .62
AB 1 2.54 2.54 .51
Error 43 214.70 4.99
Similarities A (Schools) 1 3.05 3.05 .33
B (FYHS vs. SHS) 1 4.65 4.65 .51
AB 1 .34 .34 .04Error 43 391.96 9.11
Vocabulary A (Schools) 1 1.91 1.91 .25
B (FYHS vs. SHS) 1 1.90 1.90 .25
AB 1 1.90 1.90 .25Error 43 324.89 7.55
WPPSI Verbal IQ A (Schools) 1 1.00 1.00 .01
B (FYHS vs. SHS) 1 247.00 247.00 2.19AB 1 10.00 10.00 .09Error 43 4841.00 112.55
27
TABLE 15
Means and Standard Deviations for WPPSI Performance Subtests,Performance IQ, and Full-Scale IQ (Spring 1969)
Test Grou
Animal House School 1FYHS 14
SHS 7
School 2FYHS 14
SHS 13
PictureCompletion School 1
FIRS 14SHS 7
School 2FYHS 14
SHS 13
GeometricDesigns
School 1FYHS 14
SHS 7
School 2FYHS 14
SHS 13
BlockDesigns
School 1FYHS 14
SHS 7
School 2FYHS 14
SHS 13
Mazes School 1FYHS 14
SHS 7
School 2FYHS 14SHS 13
WPPSIPerformance IQ
School 1FYHSSHS
School 2FYHSSHS
14
7
1413
WPPSIFull-Scale IQ
School 1FYHS 14
SHS 7
School 2FYHS 14
SHS 13
28
Mean SD
9.799.14
9.149.15
2.121.68
3.211.86
10.71 2.27
9.00 4.00
10.50 2.82
9.54 2.57
10.36 2.5611.14 2.97
10.21 3.2910.46 3.07
8.86 2.489.71 2.29
9.29 2.67
9.38 2.75
10.86 2.38
10.57 3.31
10.50 3.0110.38 2.63
100.71 9.59102.29 11.80
99.64 14.3498.69 13.76
90.29 10.30
91.71 11.79
89.14 13.0290.92 11.61
TABLE 16
Analysis of Covariance for WPPSI Performance Scale and WPPSI Full-Scale IQ
Test Source df Adjusted SS MS
WPPSI PerformanceSubtests
Animal House A (Schools) 1 1.55 1.55 .30
B (FYHS vs. SHS) 1 .01 .01 .00
AB '1 .84 .84 .16
Error 43 221.95 5.16
Picture A (Schools) 1 .00 .00 .00Completion B (FYHS vs. SHS) 1 8.93 8.93 1.28
AB 1 .58 .58 .08
Error 43 300.52 6.99
Geometric A (Schools) 1 10.00 10.00 1.39Designs B (FYHS vs. SHS) 1 9.94 9.94 1.38
AB 1 10.00 10.00 1.39Error 43 308.94 7.18
Block A (Schools) 1 .01 .01 .00Designs B (FYHS vs. SHS) 1 6.33 6.33 1.03
AB 1 2.08 2.08 .34Error 43 263.64 6.13
Mazes A (Schools) 1 2.31 2.31 .36B (FYHS vs. SHS) 1 .94 .94 .14AB 1 .13 .13 .02
Error 43 278.31 6.47
WPPSI A (Schools) 1. 88.00 88.00 .76Performance IQ B (FYHS vs. SHS) 1 73.00 73.00 .63
AB 1 32.00 32.00 .28Error 43 4989.00 115.99
WPPSI Full-Scale A (Schools) 1 25.00 25.00 .29IQ B (FYHS vs. SHS) 1 181.00 181.00 2.10
AB 1 1.00 1.00 .01Error 43 3702.00 86.09
29
are of a more general nature. The overall WPPSI results clearly support the
common finding of a language deficit among disadvantaged children. In view
of the limited emphasis placed on verbal skills by the kindergarten teachers,
they also suggest a need for greater stress on cognitive areas and partic-
ularly on language development in the kindergarten curriculum.
Results of PSI Gum':ookies Inventor of Factors Affectin: WPPSI Performance
end "Work" Per cent
Means and standard deviations for the four other measures comprising
the follow-up battery are summarized in Table 17. Statistical analyses of
these data are presented in Table 18. Previous IQ scores were used as a
covariate in the PSI analysis; age was the covariate for the analysis of
Gumpgookies. The mean ages in months for the School # 1 FYHS and MS groups,
and the parallel School # 2 groups were 68.75, 67.56, 69.50, and 68.38, re-
spectively. No covariate was used in the analysis of either the WPPSI
"work" 7. or Inventory of Factors Affecting Test Performance.
A statistically significant finding among this group of tests was the
higher "work" percentages earned by the children from School # 2. However,
since the "work" % measure was an experimental one in which not all exami-
ners were equally practiced, and since no other differences between schools
were found, the finding perhaps represents tester variation rather than
differences attributable to attendance at the different schools.
A trend toward a significant interaction between schools and amount of
Head Start experience is noted for both the PSI and the Inventory of Factors
Affecting Test Performance. On these measures children who attended FYHS
at School # 1 or SHS at School # 2 performed better (i.e., earned higher
PSI scores and Inventory scores) than those in the reverse situation. One
extreme score (55) on the Inventory, however, inflated the overall mean of
30
TABLE 17
Means and Standard Deviations for Four Response Measures (Spring 1969)
Test Grow N Mean SD
Preschool Inventory School 1FYHS 1.4 51.64 6.03
SHS 7 47.00 8.88
School 2FYHS 14 48.79 8.02
SHS 13 50.69 4.59
WPPSI "Work" % School 1FYHS 16 87.32 9.74
SHS 9 89.41 6.95
School 2FYHS 16 92.21 5.78
SHS 13 93.42 5.68
WPPSI Inventory of School 1Factors Affecting FYHS 16 14.56 4.23Test Performance SHS 11 18.55 13.26
School 1FYHS 17 16.53 9.71
SHS 13 12.85 2.48
Gumpgookies School 1FYHS 16 44.56 6.72
SHS 9 44.11 5.25
School 2FYHS 16 41.19 6.90
SHS I 13 45.77 3.94
31
TABLE 18
Summary of Analysis of Covariance (Preschool Inventory and Gumpgookies)and Analysis of Variance (WPPSI "Wore% and
Inventor of Factors Affectin Test Performance)
Test Source df SS MS F
Preschool Inventory A (Schools) 1 2.00 2.00 .06
B (FYHS vs. SHS) 1 2.00 2.00 .06
AB 1 102.00 102.00 3.10
Error 43 1414.00 32.88
WPPSI "Work,/ % A (Schools) 1 252.56 252.56 4.68*
B (FYHS vs. 3HS) 1 34.68 34.68 .64
AB 1 2.44 2.44 .05
Error 50 2699.94 53.99
WPPSI Inventory of A (Schools) 1 48.17 48.17 .71
Factors Affecting B (FYHS vs. SHS) 1 .31 .31 .00
Test Performance AB 1 203.25 203.25 2.99
Error 53 3608.60 68.09
Gumpgookies A (Schools) 1 11.32 11.32 .31
B (FYHS vs. SHS) 1 59.88 59.88 L.64
AB 1 80.63 80.63 2.21
Error 49 1790.12 36.53
* p< .05
32
the SHS group from School # 1.
Unfortunately, appropriate age norms for the form of the PSI used in
this study or for the Gumpgookies were not available at the time these
data were analyzed. However, comparisons of scores of the kindergarten
children with those earned by Hawaiian Head Start children participating
in the 1968-69 national evaluation may provide some further basis for
evaluating the data. The overall PSI raw score mean for the Head Start
children near the end of the program was 40.53; the PSI range for the
kindergarten groups was from 47.00 to 51.64. The overall mean Gumpgookies
raw score was 40.55 for the Head Start children; the range for kindergarten
groups was from 41.19 to 45.77. The Head Start sample made an average gain
of 13.14 points on the PSI and of 4.96 points on Gumpgookies during the
course of the year. There is perhaps no defensible basis for assuming that
raw scores increases should be constant from one age grouping to the next.
However, if one were to make this assumption, then it can be said that the
follow-up children's rate of development in kindergarten as measured by PSI
and Gumpgookies scores was somewhat slower than the rate of a group of
Hawaiian Head Start children on these same measures.
Longitudinal Analysis of WPPSI Performance IQ's
Since WPPSI Performance IQ data were obtained for three successive
years for a number of FYHS subjects and for two years for SHS subjects, it
was of interest to see what IQ changes had taken place over these time
periods. Mean WPPSI Performance IQ's for the FYHS and SHS groups for the
various testing times are summarized in Table 19. Analyses of these data
for the FYHS groups are presented in Table 20, and for the SHS groups in
Table 21. Significant differences among the IQ's for the FYHS children
from each school as well as for the combined group were obtained. That
33
TABLE 19
Mean WPPSI Performance IQ's Eor Successive Years for FYHS and SHS Children
-...
Group N
Times of Test AdministrationsFall 1967 Spring 1968 Spring 1969
FYHSSchool 1 9 89.56 93.22 97.33School 2 14 91.50 95.29 99.64Combined Schools 23 90.74 94.48 98.84
SHSSchool 1 8 -- 87.00 99.75School 2 12 -- 87.75 99.83Combined Schools 20 .. 87.45 99.80
TABLE 20
Analysis of Variance of WPPSI Performance IQ's forFYHS Children for Three Successive Years
Group Source df SS MS F
School 1 A (Subjects) 8 1468.96 183.62B (Testing Times) 2 272.52 136.26 5.11*
Linear 1 272.22 272.22 10.20**Quadratic 1 .30 .30 .01
Error 16 426.81 26.68
School 2
___
A (Subjects) 13 5417.14 416.70B (Testing Times) 2 464.90 232.45 7.46**
Linear 1 464.14 464.14 14.89**Quadratic 1 .76 .76 .02
Error 26 810.42 31.17
Combined A (Subjects) 22 6958.98 316.32 . .
Schools B (Testing Times) 2 737.04 368.52 13.10**Linear 1 736.00 736.00 26.16**Quadratic 1 1.04 1.04 .04
Error 44 1237.58 26 _3
* p < .05
** p < .03.
34
TABLE 21
Analysis of Variance of WPPSI Performance IQ's forSHS Children for Two Successive Years
Group Source d SS MS
School 1 A (Subjects) 7 1680.75 240.11B (resting Times) 1 650.25 650.25 i2.48**Error 7 364.75 52.10
School 2 A (Subjects) 11 2395.46 217.77B (resting Times) 1 876.04 876.04 6.93*Error 11 1390.45 126.40
Combined Schools A (Subjects) 19 4077.88 214.62B (Teeting Times) 1 1525.22 1525.22 16.50**Error 19 1756.28 92.44
* p <.05
** p < .01
35
these differences reflect a progressive increase in IQ over time is
indicated by the highly significant linear trend for each group analyzed
separately and when the groups are combined. Examination of the means
reveals that the overall increase IQ of approximately eight points for
the two -year period was divided about equally between the Head Start year
and the kindergarten year. The kindergarten year also included summer
Head Start attendance,but the effects of the two cannot be separated.
Examining the data for SHS children, one can see a highly significant
increase in IQ from the first testing at the start of summer Head Start
to the second testing near the end of kindergarten. Again the effects of
the summer Head Start and kindergarten programs cannot be separated. What
is especially interesting, however, is the finding that prior to any Head
Start experience, the mean IQ's for FYHS and SHS groups were very similar,
that of the FYHS children being slightly higher. However, the SHS children
gained in IQ in approximately three-fourths of a year as much as--even more
than--the FYHS children gained in twice that amount of time. Thus, at the
time of the follow-up testing, the two groups were performing at essential-
ly the same level. Secondly, assuming that the second-year gain for both
groups did not take place primarily during the summer program, it is note...
rorthy that the children's IQ's continued to increase in kindergarten. The
plateau effect that has sometimes been noted following an initial spurt in
IQ after exposure to Head Start (e.g., Wolff & Stein, 1966) clearly was not
in evidence for either the FYHS or SHS groups in this study.
36
CONCLUSIONS AND RECOMMENDATIONS
The major results of this study revealed no significant differences
between children who had attended Head Start for about a year and those who
attended a summer program only, on a group of measures administered about
eight months after the completion of Cte Head Start programs. Children
who attended one Head Start Center and subsequently enteTed kindergarten at
the associated elementary school did not perform significantly differently
from those attending another Head Start Center and elementary school. This
latter finding is not particularly surprising, since the schools were in
the same general area and served essentially the same population and no
extensive and consistent program, teacher, or other school differences were
noted. Finally, children in the FYHS and SHS groups from both schools
earned significantly higher IQ's (WPPSI Performance Scale) near the end of
kindergarten than they had prior to exposure to Head Start. The average
gain for the FYHS children was about eight points over a two-year period.
The SHS children, who had slightly lower initial IQ's, gained approximately
12 points in one year. All groups of children thus clearly benefited from
Head Start and their subsequent kindergarten experiences, although neither
length of Head Start participation nor attendance at a particular Center
or school was a significant variable.
The continued increase in IQ evidenced by the Head Start children in
kindergarten is noteworthy, since previous studies have found a leveling..
off effect following Head Start gains (Grotberg, 1969). The IQ increase
does not seem to be related to particular teacher or program variables,
since the children were distributed across 11 different kindergarten clas-
ses. The composition of the kindergarten classes of Head Start children
37
is a variable that may warrant further investigation. In this case,
Head Start children were in the minority in all classes, although their
non-Head Start peers were not from markedly different socioeconomic back-
grounds.
The failure to find significant differences between FYHS and SHS
children was consistent across all measures. Previous studies have noted
that Head Start children are rated superior by teachers in such noncogni-
tive areas as independence (Hess, 1966) and compliance with routine and
responsiveness to authority (Krider & Petsche, 1967). When kindergarten
teachers in this study were asked about their observations on the effects
of HeadStart,. comments included similar statements to the effect that
Head Start children made faster adjustments to school and assumed more
responsibility than non-Head Start children. These are important contri-
butions of Head Start, since they provide the child with attitudes and be-
haviors that enable him to learn more effectively in elementary school.
The kindergarten gains made by all Head Start children tend to support this
notion of the impact of Head Start. The failure to find significant dif-
ferences between FYHS and SHS children raises the question, however, of
whether a full year is necessary for the development of some of these
"school-adaptive" characteristics. If the primary goal of Head Start is
to enable the child to function more effectively in elementary school by
teaching him appropriate school behaviors, then a summer program may be
able to accomplish this as effectively as a full-year program. On the other
hand, if Head Start is also to provide the child with significantly better
cognitive skills, more extensive training would seem to be necessary. It
is suggested that FYHS programs with defined curricula focused on cognitive
development are more likely to result in differential performance between
38
children exposed to these curricula and those attending summer Head Start
or no Head Start program at all. It is recommended, therefore, that future
follow-up studies concentrate primarily on the long-range effects of such
MS programs.
39
APPENDIX
Inter-correlations Among Follow-up Measures andPrevious Intelligence Tf2st Data
1.
StanfordeBinet
(Fall 1967)
2.
Stanford-Binet
1;88
(Spring 1968)
3.
WPPSI Performance IQ
,56 1
(Fall 1967)
4.
WPPSI Performance IQ
.63
(Spring and Summer
1968)
5.
WPPSI Performance IQ
.42
a(Spring 1969)
6.
WPPSI Verbal IQ
J..74
(Spring 1969)
7.
WPPSI Full-Scale IQ
.67
1(Spring 1969)
8.
WPPSI Inventory of
T:02
Factors Affecting
Test Performancea
(Spring 1969)
9.
Preschool Inventory
.54
(Spring 1969)
23
46
78
9
.88**
.56*
.63**
.42*
.74**
.67**
-.02
.54**
(19)
(16)
(22)
(24)
(24)
(24)
(24)
(24)
.50*
.40*
.59**
.70**
.72**
-.28
.75**
(21)
(35)
(36)
(34)
(34)
.(3-6)
(36)
.50
.77**
.79**
.55**
.77**
-.22
.62**
(23)
(23)
(23)
(23)
(23)
(23)
.40
.77
.53**
.44**
.58**
.02
.42**
(50)
(49)
(49)
(50)
(50)
.59
.79
.53
la O
P.44**
.83**
-.36**
.59**
(54)
(54)
(57)
(57)
.70
.55
.44
.44
Q. P
O.87**
-.27*
.76**
(54)
(54)
(54)
.72
.77
.58
.83
t.87
WO
OP
-.15
.71**
(54)
(54)
-.28
-.22
-.02
-.36
1-.27
-.15
op O
M-.64**
(57)
.75
.62
.42
.59
t.76
.71
-.64
OP
pa.
10
I.05
(24)
.45**
(36)
.24
_(23)
. 17
(50)
.39**
(57)
.27*
(54)
.33*
(54)
-..45**
(57)
.51**
(57)
APPENDIX (Cont.)
25
79
10.
Gumpgookies
(Spring 1969)
.05
.45
.24
.17
.39
.27
.33
-.45
.51
Note.--Sample sizes are in parentheses.
aLow scores reflect more adaptive test behavior.
* p<.05
** p<.01
REFERENCES
Grotberg, E. H. Review of research 1965 to 1969. Research and EvaluationOffice, Project Head Start, Office of Economic Opportunity, June 1969.
Hertzig, M. E.; Birch, H. G.; Thomas, A.; & Mendez, O. A. Class andethnic differences in the responsiveness of preschool children tocognitive demands. Monographs of the Society for Research inChild Development, 1968, Serial No. 117.
Hess, R. Techniques for assessing cognitive and social abilities ofchildren and parents in project Head Start. Research in Education(ED 015 772), 1966.
Krider, M. A. and Petsche, M. An evaluation of Head Start preschoolenrichment programs as they affect the intellectual ability, thesocial adjustment, and achievement time of five-year-old childrenenrolled in Lincoln, Nebraska. Research in Education (ED 015 011),1967.
Wolff, M. and Stein, A. Six months later, a comparison of children whohad Head Start summer, 1965, with their classmates in kindergarten.Research in Education (ED 015 025), 1966.
42